Auto insurance underwriting is the process by which insurance companies determine whether to offer insurance coverage, whether to renew insurance coverage, and to determine the pricing of any coverage that is offered. Insurance pricing may be based on a rate which may then be adjusted based on discounts, credits, penalties and other adjustments. The final premium may be based on the determined risk factors associated with the driver, vehicle, laws/regulations, and other business factors.
Insurance pricing is typically derived using correlative data as a proxy for driving behavior. The proxies, such as age, driving experience, occupation, etc. may be derived from actuarial determined data. The pricing can vary depending on many factors that are believed to have an impact on the expectation of future claims and any cost associated with such future claims.
Generally, the three major factors in assessing risk may be: 1) coverage; 2) vehicle; and 3) driver. The coverage may determine the type and amount for which the insurance company may be responsible. The vehicle and driver may be important based on the statistical data that may indicate that a college educated professional driving a Lamborghini may not pose the same risk as a male high school student driving a station wagon.
In determining the pricing, the insurance company may determine the vehicle and coverage with some level of certainty. For example, the insurance company is provided with the vehicle manufacturer, model, age, value (and possibly service history) for which coverage is being requested. The insurance company may also determine the type of coverage, (e.g. liability, collision, comprehensive, personal injury protection, and uninsured motorist protection), that is being purchased.
However, the insurance company may have little data for identifying the amount of time a vehicle is being operated by a particular driver. For example, in a household with multiple drivers and multiple vehicles, neither the insurance company nor the customer may possess accurate information regarding amount of time each vehicle is used by a particular individual. Further, those individuals are assessed based on correlative data, but this may not be accurate, e.g., not all high school students drive in a similar manner.
Accordingly, methods and apparatus are described in greater detail that use telematics data to identify driver signatures associated with the use of the vehicle. The system may further be configured to identify the driver associated with the driver signatures. This may allow the insurance company to determine risk associated with offering coverage and allow the insurance company to adjust pricing to reflect the actual usage of a particular vehicle. The methods described herein may use passive and non-passive techniques to identify a driver signature associated with use of the vehicle and a driver associated with each driver signature.